* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Storing Data: Disks and Files - Department of Computer Science
Survey
Document related concepts
Transcript
Storing Data: Disks and Files Chapter 7 Jianping Fan Dept of Computer Science UNC-Charlotte Database Management Systems, R. Ramakrishnan and J. Gehrke 1 Three Materials for Database Storage Main Memory (RAM or Buffer): we use it to store currently-accessing data tuples or tables and their joins ---data will disappear when we turn off computers Hard Disks: we use them to store database tables and indexing structures Data Tapes: we may not use tapes nowadays Every tuples to be accessed should be available in buffer! Database Management Systems, R. Ramakrishnan and J. Gehrke 2 Disks and Files The tuples to be accessed should be in RAM! DBMS stores data on (“hard”) disks. This has major implications for DBMS design! – READ: transfer data from disk to main memory (RAM). – WRITE: transfer data from RAM to disk. I/O cost – Both are high-cost operations, relative to in-memory operations, so must be planned carefully! Database Management Systems, R. Ramakrishnan and J. Gehrke 3 Why Not Store Everything in Main Memory? It costs much more. $65-200 will buy you either 24GB of RAM or 300GB of disk today. Main memory is volatile. We want data to be saved between runs. (Obviously! When we turn off computers, data in RAM will lost) Storage hierarchy: – – – Main memory (RAM) for currently used data. Disk for the main database (secondary storage). Tapes for archiving older versions of the data (third storage). We may not use frequently now. Database Management Systems, R. Ramakrishnan and J. Gehrke 4 Disks Secondary storage device of choice. Main advantage over tapes: random access vs. sequential. Data is stored and retrieved in units called disk blocks or pages. Unlike RAM, time to retrieve a disk page varies depending upon location on disk. – Therefore, relative placement of pages on disk has major impact on DBMS performance! Storage management is a critical issue for DBMS! Database Management Systems, R. Ramakrishnan and J. Gehrke 5 Components of a Disk Disk head Spindle Tracks The platters rotate. The arm assembly is moved in or out to position a head on a desired track. Tracks under heads make a cylinder (imaginary!). Sector Arm movement Only one head reads/writes at any one time. Platters Arm assembly Block size is a multiple of sector size (which is fixed). Database Management Systems, R. Ramakrishnan and J. Gehrke 6 Accessing a Disk Page Time to access (read/write) a disk block: – – – Seek time and rotational delay dominate. – – – seek time (moving arms from disk head to the requested track) rotational delay (waiting for requested block to be rotated under head) transfer time (moving data from disk to main memory) Seek time varies from about 1 to 20 millisecond (msec) Rotational delay varies from 0 to 10msec Transfer rate is about 1msec per 4KB page Key to lower I/O cost: reduce seek/rotation delays! Hardware vs. software solutions? Database Management Systems, R. Ramakrishnan and J. Gehrke 7 Arranging Pages on Disk `Next’ block concept: – – – blocks on same track, followed by blocks on same cylinder, followed by blocks on adjacent cylinder Blocks in a file should be arranged sequentially on disk (by `next’), to minimize seek and rotational delay. For a sequential scan, pre-fetching several pages at a time is a big win! Database Management Systems, R. Ramakrishnan and J. Gehrke 8 RAID (Redundant Arrays of Independent Disks) R S Disk Array: Arrangement of several disks that gives abstraction of a single, large disk. Goals: Increase performance and reliability. Database Management Systems, R. Ramakrishnan and J. Gehrke 9 Data Stripping: Redundancy: Mirrored disks Two main techniques: (a) Data Striping: Data is partitioned; size of a partition is called the striping unit. Partitions are distributed over several disks. (b) Redundancy: More disks -> more failures. Redundant information allows reconstruction of data if a disk fails. Database Management Systems, R. Ramakrishnan and J. Gehrke 10 Single Disk Redundancy: Mirrored disks Benefits? Database Management Systems, R. Ramakrishnan and J. Gehrke 11 RAID Levels Level 0: No redundancy Level 1: Mirrored (two identical copies) – – – Each disk has a mirror image (check disk) Parallel reads, a write involves two disks. Maximum transfer rate = transfer rate of one disk Database Management Systems, R. Ramakrishnan and J. Gehrke 12 RAID Levels Level 0+1: Striping and Mirroring – Parallel reads, a write involves two disks. – Maximum transfer rate = aggregate bandwidth Database Management Systems, R. Ramakrishnan and J. Gehrke 13 Disk Striping: overlapping of disk reads and writes Controller channel channel 0 8 1 9 2 10 3 11 4 12 Database Management Systems, R. Ramakrishnan and J. Gehrke 5 13 6 14 7 15 14 Disk Mirroring: read can be speeded up! Controller channel channel 0 1 2 3 0’ 1’ 2’ 3’ 4 5 6 7 4’ 5’ 6’ 7’ Database Management Systems, R. Ramakrishnan and J. Gehrke 15 RAID Levels (Contd.) Level 3: Bit-Interleaved Parity – – Striping Unit: One bit. One check disk. Each read and write request involves all disks; disk array can process one request at a time. Reed-Solomon Code What is error-correction code? Database Management Systems, R. Ramakrishnan and J. Gehrke 16 RAID Levels (Contd.) Level 4: Block-Interleaved Parity – Striping Unit: One disk block. One check disk. – Parallel reads possible for small requests, large requests can utilize full bandwidth – Writes involve modified block and check disk Total bits 7168, each block 1024, store in 1 disk Database Management Systems, R. Ramakrishnan and J. Gehrke 17 RAID Levels (Contd.) Level 4: Block-Interleaved Distributed Parity – Similar to RAID Level 3, but parity blocks are distributed over all disks Total bits 7168, each block 1024, store in 4 disks Database Management Systems, R. Ramakrishnan and J. Gehrke 18 Disk Space Management Lowest layer of DBMS software manages space on disk. Higher levels call upon this layer to: – – allocate/de-allocate a page read/write a page Request for a sequence of pages must be satisfied by allocating the pages sequentially on disk! Higher levels don’t need to know how this is done, or how free space is managed. Database Management Systems, R. Ramakrishnan and J. Gehrke 19 Everything to be accessed should be in buffer! Buffer Management in a DBMS Page Requests from Higher Levels BUFFER POOL disk page free frame MAIN MEMORY DISK DB choice of frame dictated by replacement policy Data must be in RAM for DBMS to operate on it! Table of <frame#, pageid> pairs is maintained. Database Management Systems, R. Ramakrishnan and J. Gehrke 20 Buffer Management in a DBMS Frame: basic unit for buffer management Frame is described by two parameters: (a) pin_count: how many users are currently using the frame (b) dirty: the current frame is modified or not? on or turned off Initially, dirty is set as turned off, and pin_count is set as 0. Database Management Systems, R. Ramakrishnan and J. Gehrke 21 When a Page is Requested ... If requested page is in pool: - Increase the pin_count for the requested frame Pin the page and return its address. If requested page is not in pool: – – – Choose a frame for replacement If the selected frame is dirty, write it to disk Read requested page into chosen frame Pin the page and return its address. Database Management Systems, R. Ramakrishnan and J. Gehrke 22 How to select the replace frame ... If the requested page is not in the buffer pool, and but the buffer pool has free frame ---transform the requested page from disk to the free frame If the requested page is not in the buffer pool, and if the buffer pool does not have the available free frame ----select the frame with pin_count == 0 for replacement If requests can be predicted (e.g., sequential scans) pages can be pre-fetched several pages at a time! Database Management Systems, R. Ramakrishnan and J. Gehrke 23 More on Buffer Management Requestor of page must unpin it, and indicate whether page has been modified: – Page in pool may be requested many times, – dirty bit is used for this. a pin count is used. A page is a candidate for replacement iff pin count = 0. CC & recovery may entail additional I/O when a frame is chosen for replacement. (Write-Ahead Log protocol; more later.) Database Management Systems, R. Ramakrishnan and J. Gehrke 24 Buffer Replacement Policy Frame is chosen for replacement by a replacement policy: – Least-recently-used (LRU), Clock, MRU etc. Policy can have big impact on # of I/O’s; depends on the access pattern. Sequential flooding: Nasty situation caused by LRU + repeated sequential scans. – # buffer frames < # pages in file means each page request causes an I/O. MRU much better in this situation (but not in all situations, of course). Database Management Systems, R. Ramakrishnan and J. Gehrke 25 DBMS vs. OS File System OS does disk space & buffer mgmt: why not let OS manage these tasks? Differences in OS support: portability issues Some limitations, e.g., files can’t span disks. Buffer management in DBMS requires ability to: – – pin a page in buffer pool, force a page to disk (important for implementing CC & recovery), adjust replacement policy, and pre-fetch pages based on access patterns in typical DB operations. Database Management Systems, R. Ramakrishnan and J. Gehrke 26 Record Formats: Fixed Length F1 F2 F3 F4 L1 L2 L3 L4 Base address (B) Address = B+L1+L2 Information about field types same for all records in a file; stored in system catalogs. Finding i’th field requires scan of record. Database Management Systems, R. Ramakrishnan and J. Gehrke 27 Record Formats: Variable Length Two alternative formats (# fields is fixed): F1 4 Field Count F2 $ F3 $ F4 $ $ Fields Delimited by Special Symbols F1 F2 F3 F4 Array of Field Offsets Second offers direct access to i’th field, efficient storage of nulls (special don’t know value); small directory overhead. Database Management Systems, R. Ramakrishnan and J. Gehrke 28 Page Formats: Fixed Length Records Slot 1 Slot 2 Slot 1 Slot 2 Free Space ... Slot N ... Slot N Slot M N PACKED 1 . . . 0 1 1M number of records M ... 3 2 1 UNPACKED, BITMAP number of slots Record id = <page id, slot #>. In first alternative, moving records for free space management changes rid; may not be acceptable. Database Management Systems, R. Ramakrishnan and J. Gehrke 29 Page Formats: Variable Length Records Rid = (i,N) Page i Rid = (i,2) Rid = (i,1) 20 N ... 16 2 24 N 1 # slots SLOT DIRECTORY Pointer to start of free space Can move records on page without changing rid; so, attractive for fixed-length records too. Database Management Systems, R. Ramakrishnan and J. Gehrke 30 Files of Records Page or block is OK when doing I/O, but higher levels of DBMS operate on records, and files of records. FILE: A collection of pages, each containing a collection of records. Must support: – – – insert/delete/modify record read a particular record (specified using record id) scan all records (possibly with some conditions on the records to be retrieved) Database Management Systems, R. Ramakrishnan and J. Gehrke 31 Unordered (Heap) Files Simplest file structure contains records in no particular order. As file grows and shrinks, disk pages are allocated and de-allocated. To support record level operations, we must: – – – keep track of the pages in a file keep track of free space on pages keep track of the records on a page There are many alternatives for keeping track of this. Database Management Systems, R. Ramakrishnan and J. Gehrke 32 Heap File Implemented as a List Data Page Data Page Data Page Full Pages Header Page Data Page Data Page Data Page Pages with Free Space The header page id and Heap file name must be stored someplace. Each page contains 2 `pointers’ plus data. Database Management Systems, R. Ramakrishnan and J. Gehrke 33 Heap File Using a Page Directory Data Page 1 Header Page Data Page 2 DIRECTORY Data Page N The entry for a page can include the number of free bytes on the page. The directory is a collection of pages; linked list implementation is just one alternative. – Much smaller than linked list of all HF pages! Database Management Systems, R. Ramakrishnan and J. Gehrke 34 Indexes A Heap file allows us to retrieve records: – – Sometimes, we want to retrieve records by specifying the values in one or more fields, e.g., – – by specifying the rid, or by scanning all records sequentially Find all students in the “CS” department Find all students with a gpa > 3 Indexes are file structures that enable us to answer such value-based queries efficiently. Database Management Systems, R. Ramakrishnan and J. Gehrke 35 System Catalogs For each index: – For each relation: – – – – name, file name, file structure (e.g., Heap file) attribute name and type, for each attribute index name, for each index integrity constraints For each view: – structure (e.g., B+ tree) and search key fields view name and definition Plus statistics, authorization, buffer pool size, etc. Catalogs are themselves stored as relations! Database Management Systems, R. Ramakrishnan and J. Gehrke 36 Attr_Cat(attr_name, rel_name, type, position) attr_name attr_name rel_name type position sid name login age gpa fid fname sal rel_name Attribute_Cat Attribute_Cat Attribute_Cat Attribute_Cat Students Students Students Students Students Faculty Faculty Faculty Database Management Systems, R. Ramakrishnan and J. Gehrke type string string string integer string string string integer real string string real position 1 2 3 4 1 2 3 4 5 1 2 3 37 Summary Disks provide cheap, non-volatile storage. – Random access, but cost depends on location of page on disk; important to arrange data sequentially to minimize seek and rotation delays. Buffer manager brings pages into RAM. – – – – Page stays in RAM until released by requestor. Written to disk when frame chosen for replacement (which is sometime after requestor releases the page). Choice of frame to replace based on replacement policy. Tries to pre-fetch several pages at a time. Database Management Systems, R. Ramakrishnan and J. Gehrke 38 Summary (Contd.) DBMS vs. OS File Support – DBMS needs features not found in many OS’s, e.g., forcing a page to disk, controlling the order of page writes to disk, files spanning disks, ability to control pre-fetching and page replacement policy based on predictable access patterns, etc. Variable length record format with field offset directory offers support for direct access to i’th field and null values. Slotted page format supports variable length records and allows records to move on page. Database Management Systems, R. Ramakrishnan and J. Gehrke 39 Summary (Contd.) File layer keeps track of pages in a file, and supports abstraction of a collection of records. – Pages with free space identified using linked list or directory structure (similar to how pages in file are kept track of). Indexes support efficient retrieval of records based on the values in some fields. Catalog relations store information about relations, indexes and views. (Information that is common to all records in a given collection.) Database Management Systems, R. Ramakrishnan and J. Gehrke 40